© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 12587
    Prev
    Next
    INDEX
    Explanations

    legal citations and case names

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on the abbreviated names of law‐report series (e.g. the “P” in “P.2d” or the “S” in “S.E.2d”), i.e. reporter abbreviations in legal citations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⚌
    -1.70
    daun
    -1.63
    href
    -1.59
    _
    -1.59
     “
    -1.58
    <blockquote>
    -1.56
    IDAS
    -1.52
    ྃ
    -1.51
    allclose
    -1.49
    ’
    -1.48
    POSITIVE LOGITS
    ícias
    1.54
    但是在
    1.36
    很难
    1.30
     ながら
    1.27
    pography
    1.27
    ⦊
    1.26
     varandra
    1.26
    s
    1.25
     it
    1.24
    ykjav
    1.24
    Activations Density 0.028%

    No Known Activations